In the latest OCAD Update, the option Create Feature Map was added to the DEM Wizard to gain even more information from LiDAR data.
The Feature Map can be used to recognize objects close to the ground such as stones, walls, tree trunks, fences, or cars which were previously not or only poorly recognizable on background maps that can be created with the DEM Wizard in OCAD.
How the creation of Feature Maps works
In a LiDAR file, each point is typically assigned to a class based on the reflection of the laser pulse, such as ground, vegetation, buildings or water. Only ground points are used to generate the Hill Shading and Slope Gradient map. This means that a lot of information is lost that is available in points classified differently.
To generate the Feature Map, OCAD reclassifies the LiDAR points so that not only ground points but all points are used for the calculation. You can further define from and up to which height points are taken into account.
Example stone detection
Pontresina, Switzerland
LiDAR data from 2022, 29 points per square meter, 0.5m cell size, 0.0-2.0m threshold.
Example stone wall detection
Jura, Switzerland
LiDAR data from 2020, 25 points per square meter, 0.5m cell size, 0.0-1.5m threshold.
Example tree trunks detection
S-chanf, Switzerland
LiDAR data from 2022, 25 points per square meter, 0.5m cell size, 0.0-0.5m threshold.
Example urban area
Zeiningen, Switzerland
LiDAR data from 2020, 13 points per square meter, 0.5m cell size, 0.0-2.0m threshold.
Example without satisfactory result
Lillehammer, Norway
LiDAR data from 2017, 10 points per square meter, 0.75m cell size, 0.0-2.0m threshold.
Conclusion:
The information content of the feature map depends on the terrain type, the settings you choose in the dialog and the quality of the LiDAR data, in particular the point density. With good data quality, the feature map can be a useful addition to the existing background maps, to detect objects and help the cartographer to determine the exact position in the terrain.
Credit goes to Jeff Teutsch and his Lidar Case Study – Using simple ground reclassification to see features in data.